AI Image Noise Filtering for Ghosting and Blur Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise reduction technologies for images fail to consider the intensity of side effects caused by noise filters, leading to issues like blurring, resolution degradation, ghosting, and motion blur, especially in stationary and moving objects.
Innovation Solution
An image noise reduction device and method using machine learning to measure side effects caused by noise filters, adjusting filter intensity based on pixel differences and high-frequency component analysis, and switching between 2D and 3D noise filters to minimize image quality degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 3D noise filter is applied to remove noise in stationary objects, then noise removal performance is improved, but ghost effect and motion blur occur in moving subjects
Solution Approach 1:
The patent applies dynamics by making the noise filter intensity adjustable rather than fixed. The system dynamically adjusts the noise filter intensity based on real-time analysis of image characteristics, switching between strong filtering for stationary objects and weak filtering for moving objects, thereby adapting to different scene requirements and avoiding ghost effects in moving subjects
Solution Approach 2:
The patent applies local quality by differentiating the noise filtering strength across different regions of the image. Through motion detection and region analysis, the system applies strong noise filtering to stationary background regions while using weak or no filtering in moving object regions, allowing each area to receive appropriate processing intensity
2Reliability
If 2D noise filter is used to correct noise in moving objects, then noise reduction is effective for moving subjects, but blurring and resolution degradation occur in stationary objects
Solution Approach 1:
The system dynamically switches between 2D and 3D noise filtering modes based on motion detection. When moving objects are detected, 2D filtering is applied to preserve motion clarity; when stationary objects are identified, 3D filtering is activated to enhance noise removal and sharpness, thereby optimizing performance for each object type
Solution Approach 2:
The patent applies local quality by selectively applying different filtering algorithms to different spatial regions. Moving objects receive 2D filtering treatment while stationary objects receive 3D filtering, allowing each region to benefit from the most appropriate noise reduction technique without compromising overall image quality
3Reliability
If noise filter intensity is increased to remove more noise, then noise removal effectiveness is improved, but side effects such as blurring and resolution degradation become more severe
Solution Approach 1:
The patent implements feedback by continuously analyzing image characteristics and motion information to adjust noise filter intensity in real-time. The system monitors the balance between noise removal and quality preservation, reducing filter intensity when side effects are detected and increasing it when noise levels require stronger filtering, thereby maintaining optimal performance
Solution Approach 2:
The system changes the noise filter intensity parameter dynamically based on scene analysis. By adjusting the filtering strength parameter in response to detected motion and image characteristics, the system optimizes the trade-off between noise removal effectiveness and quality preservation, preventing excessive blurring while maintaining sharpness
Data Source
AI summary
An image noise learning server includes an image input interface configured to receive training images, and at least one processor configured to control an image extractor to extract, from the training images, a first image including a stationary object and a second image including a moving object, a noise filter to obtain a third image by applying noise filtering with a first intensity to the second image, the third image including the moving object, a labeling unit to determine an intensity of a side effect based on a difference between the stationary object included in the first image and the moving object included in the third image, and a machine learning unit to receive, as a label, the determined intensity of the side effect and image attributes of the training images, and obtain artificial intelligence (AI) parameters by performing machine learning on the second image based on the received label.